9430738

Automated Emotional Clustering of Social Media Conversations

PublishedAugust 30, 2016
Assigneenot available in USPTO data we have
Technical Abstract

Patent Claims
35 claims

Legal claims defining the scope of protection, as filed with the USPTO.

1

1. A computer-implemented real-time, emotional-monitoring system for brands and companies, comprising a computer system comprising hardware and software, said computer system implementing the steps of obtaining real-time social media conversations, inputting said real-time social media conversations into the computer system, wherein the software automatically classifies, categorizes and summarizes expressed emotions in said real-time social media conversations by using a knowledge base of emotional words, or phrases, or both words and phrases as an input, wherein the knowledge base comprises a list of emotional tokens and pairwise token distance using a degree of similarity, each pair associated with a valence and arousal score expressed on a scale, defining a continuous distance metric between said real-time social media conversations comprising same or different number of the emotional tokens utilizing an expand-align-compare (EAC) algorithm, conducting hierarchical clustering based on that distance metric, and outputting a result associated with said hierarchical clustering.

2

2. A computer-implemented method for real-time, emotional-monitoring system for brands and companies, comprising the steps of obtaining real-time social media conversations, inputting said real-time social media conversations into a computer system comprising software that automatically classifies, categorizes and summarizes expressed emotions in said real-time social media conversations by using a knowledge base of emotional words, or phrases, or both words and phrases as an input, wherein the knowledge base comprises a list of emotional tokens and pairwise token distance using a degree of similarity, each pair associated with a valence and arousal score expressed on a scale, defining a continuous distance metric between said real-time social media conversations comprising same or different number of the emotional tokens utilizing an expand-align-compare (EAC) algorithm, conducting hierarchical clustering based on that distance metric, and outputting a result associated with said hierarchical clustering.

3

3. Non-transitory computer readable media storing computer code that, when executed, performs the computer-implemented method of claim 2 .

4

4. A social media intelligence platform comprising: a. a processor providing an analyst application comprising: i. a first software module obtaining real-time social media conversations; ii. a second software module automatically classifying, categorizing and summarizing expressed emotions in said real-time social media conversations by using a knowledge base of emotional words, or phrases, or both words and phrases as an input, wherein the knowledge base comprises a list of emotional tokens and pairwise token distance using a degree of similarity, each pair associated with a valence and arousal score expressed on a scale; iii. a third software module defining a continuous distance metric between said real-time social media conversations comprising same or different number of the emotional tokens utilizing an expand-align-compare (EAC) algorithm; and iv. a fourth software module conducting clustering based on that distance metric; b. the processor further providing a customer dashboard application comprising: i. a fifth software module providing an interface allowing the customer to input a topic; and ii. a sixth software module providing a graphic visualization of the sentiment clusters for the topic.

5

5. The platform of claim 4 , wherein the social media comprises a blog, microblog, social network, podcast, wiki, content community, virtual world, or a combination thereof.

6

6. The platform of claim 4 , wherein the knowledge base is domain-specific.

7

7. The platform of claim 6 , wherein the domain is entertainment.

8

8. The platform of claim 4 , wherein the knowledge base further comprises words or phrases not expressing sentiment.

9

9. The platform of claim 4 , wherein the analyst application further comprises a software module associating equivalents with at least one of the tokens.

10

10. The platform of claim 9 , wherein the equivalents comprise synonyms or misspellings.

11

11. The platform of claim 4 , wherein the analyst application further comprises a software module calibrating the tokens by assigning a valence and arousal score.

12

12. The platform of claim 11 , wherein the valence and arousal scores are derived from a sample of social media communications containing at least one of the tokens.

13

13. The platform of claim 4 , wherein the number of sentiment clusters is dynamically determined based on the content of the social media communications.

14

14. The platform of claim 4 , wherein the graphic visualization of sentiment clusters comprises one or more prototypical communications associated with each cluster.

15

15. The platform of claim 4 , wherein the customer dashboard application is offered as software-as-a-service.

16

16. A computer-implemented social media intelligence system comprising: a. a digital processor performing a computer program comprising executable instructions stored on a memory device; b. the computer program providing a social media intelligence application comprising: i. a first software module obtaining real-time social media conversations; ii. a second software module automatically classifying, categorizing and summarizing expressed emotions in said real-time social media conversations by using a knowledge base of emotional words, or phrases, or both words and phrases as an input, wherein the knowledge base comprises a list of emotional tokens and pairwise token distance using a degree of similarity, each pair associated with a valence and arousal score expressed on a scale; iii. a third software module defining a continuous distance metric between said real-time social media conversations comprising same or different number of the emotional tokens utilizing an expand-align-compare (EAC) algorithm; iv. a fourth software module conducting hierarchical clustering based on that distance metric; and vi. a fifth software module outputting a result associated with said hierarchical clustering.

17

17. The system of claim 16 , wherein the social media comprises a blog, microblog, social network, podcast, wiki, content community, virtual world, or a combination thereof.

18

18. The system of claim 16 , wherein the knowledge base is domain-specific.

19

19. The system of claim 18 , wherein the domain is entertainment.

20

20. The system of claim 16 , wherein the knowledge base further comprises words or phrases not expressing sentiment.

21

21. The system of claim 16 , wherein at least one of the tokens is associated with at least one equivalent.

22

22. The system of claim 21 , wherein an equivalent comprises a synonym or a misspelling.

23

23. The system of claim 16 , wherein the valence and arousal scores are derived from a sample of social media communications containing at least one of the tokens.

24

24. The system of claim 16 , wherein the number of clusters is dynamically determined based on the content of the social media communications.

25

25. The system of claim 16 , wherein the result comprises a graphic visualization of the clusters.

26

26. Non-transitory computer-readable storage media encoded with a computer program including instructions executable by a processor to provide a social media intelligence application comprising: a. a first software module obtaining real-time social media conversations; b. a second software module automatically classifying, categorizing and summarizing expressed emotions in said real-time social media conversations by using a knowledge base of emotional words, or phrases, or both words and phrases as an input, wherein the knowledge base comprises a list of emotional tokens and pairwise token distance using a degree of similarity, each pair associated with a valence and arousal score expressed on a scale; c. a third software module defining a continuous distance metric between said real-time social media conversations comprising same or different number of the emotional tokens utilizing an expand-align-compare (EAC) algorithm; d. a fourth software module conducting hierarchical clustering based on that distance metric; and e. a fifth software module outputting a result associated with said hierarchical clustering.

27

27. The media of claim 26 , wherein the social media comprises a blog, microblog, social network, podcast, wiki, content community, virtual world, or a combination thereof.

28

28. The media of claim 26 , wherein the knowledge base is domain-specific.

29

29. The media of claim 28 , wherein the domain is entertainment.

30

30. The media of claim 26 , wherein the knowledge base further comprises words or phrases not expressing sentiment.

31

31. The media of claim 26 , wherein at least one of the tokens is associated with at least one equivalent.

32

32. The media of claim 31 , wherein an equivalent comprises a synonym or a misspelling.

33

33. The media of claim 26 , wherein the valence and arousal scores are derived from a sample of social media communications containing at least one of the tokens.

34

34. The media of claim 26 , wherein the number of clusters is dynamically determined based on the content of the social media communications.

35

35. The media of claim 26 , wherein the result comprises a graphic visualization of the clusters.

Patent Metadata

Filing Date

Unknown

Publication Date

August 30, 2016

Inventors

Ka-Chuen Sam Hui
Jared Adam Feldman

Want to explore more patents?

Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.

Citation & reuse

Analysis on this page is generated by Patentable — an AI-powered patent intelligence platform. AI-generated summaries, explanations, and analysis may be reused with attribution and a visible link back to the canonical URL below. Patent abstracts and claims are USPTO public domain.

Cite as: Patentable. “AUTOMATED EMOTIONAL CLUSTERING OF SOCIAL MEDIA CONVERSATIONS” (9430738). https://patentable.app/patents/9430738

© 2026 Patentable. All rights reserved.

Patentable is a research and drafting-assistant tool, not a law firm, and does not provide legal advice. Documents we generate are drafts for review by a licensed patent attorney.